海洋测绘2024,Vol.44Issue(6) :44-48.DOI:10.3969/j.issn.1671-3044.2024.06.009

基于浅剖声纳的海底浅地层声学空白带探测

Acoustic blank zone detection from sub-bottom profiler sonar data

李正元 李邵波 贺惠忠 陈婷婷
海洋测绘2024,Vol.44Issue(6) :44-48.DOI:10.3969/j.issn.1671-3044.2024.06.009

基于浅剖声纳的海底浅地层声学空白带探测

Acoustic blank zone detection from sub-bottom profiler sonar data

李正元 1李邵波 2贺惠忠 1陈婷婷1
扫码查看

作者信息

  • 1. 自然资源部 海洋环境探测技术与应用重点实验室,广东 广州 510300;国家海洋局 南海调查技术中心,广东 广州 510300
  • 2. 自然资源部 海洋环境探测技术与应用重点实验室,广东 广州 510300;中国地质大学(武汉)地理与信息工程学院,湖北 武汉 430074
  • 折叠

摘要

当前基于浅地层剖面仪(浅剖)数据的声学空白带探测方法缺失,人工探测提取费时费力且依赖于人工经验.针对这一现状提出了两种声学空白带探测方法,分别适用于浅层、深层空白带探测.首先基于浅层声学空白带导致的下部地层回波能量弱、直方图分布为截尾分布这一特征,给出了基于直方图分布特征的声学空白带探测方法.其次针对深部声学空白带无明显声阻抗差异界面特点,通过层界提取算法先获取地层结构信息,再建立下部地层结构起伏变化函数模型,基于函数极值点判断给出声学空白带位置.实验表明提出的方法探测精确率均优于 80%、召回率优于 90%,实现了可靠的声学空白带探测.

Abstract

At present,the acoustic blank zone detection method based on sub-bottom(SBP)is lacking,and manual detection and extraction is time-consuming and laborious and depends on manual experience.In view of this situation,two acoustic blank zone detection methods are proposed,which are suitable for shallow and deep blank zone detection respectively.Firstly,based on the characteristics of weak echo energy and truncated histogram distribution caused by shallow acoustic blank zone,an acoustic blank zone detection method based on histogram distribution characteristics is presented.Secondly,in view of the interface characteristic of the deep acoustic blank zone without obvious acoustic impedance difference,the formation structure information is obtained by using the horizon picking algorithm,and then the fluctuation function model of the lower formation structure is established,and the position of the acoustic blank zone is determined by the extreme point of the fitted function.The experimental results show that the detection accuracy and recall rate of the proposed method are better than 80%and 90%respectively,thus achieving reliable acoustic blank band detection.

关键词

声学空白带探测/浅地层剖面图像/特征提取/地层结构提取/层界提取

Key words

acoustic blank zone detection/sub-bottom profile image/feature extraction/reflector structure extraction/horizon picking

引用本文复制引用

出版年

2024
海洋测绘
海军海洋测绘研究所

海洋测绘

CSTPCD北大核心
影响因子:0.669
ISSN:1671-3044
段落导航相关论文